Matt, I thought that was the case for Annex M. More of the lower frequencies allocated to upstream, therefore the downstream starts later. Although I thought for Annex A it just put more bits in the same frequency bins to get you from c400kps to c800kps, a bit like it does on the downstream ADSL1 Vs ADSL2. Regards, David
In Annex A ('Normal' UK ADSL/ADSL2/ADSL2+), upstream uses frequencies from 26kHz to 137.825kHz, downstream uses 138kHz upwards (Annex M looses the frequencies from 138kHz to 276kHz from downstream to provide the extra bins for the increased upstream rates).
Call me 'w23' At any given moment in the universe many things happen. Coincidence is a matter of how close these events are in space, time and relationship. Opinions expressed in forum posts are those of the poster, others may have different views.
Ah, there we go - I'm confusing annex A and M. Sorry about that. Right, just asked further (hopefully I've not got it wrong this time), and basically when the upstream is capped there's a gap between the up and downstream frequencies that helps absorb any errors with the frequencies bleeding into each other. With the whole upstream frequency allocation in use there's much less margin for error there, which can lead to issues.
Matt, Thanks. I made the same mistake last week, got my Annex mixed up with my ADSL - hence the question. Once I am out of my new 10 day training period I may give it a go. Thanks, David
Niko, According to Wikipedia, Phone uses lowest frequencies, then a gap, then Upstream, then small gap, then downstream. What we need is someone to test this who has a modem capable of showing frequency bins, bit loading etc - DMT tool or Routerstats
On a BT business hub (2 Wire HGV2701) in my office, it is running ADSL2 (not ADSL2+) with uncapped upstream, typical upstream sync around 1000kbps, this currently has a 'gap' of 2 tones between upstream (uses tones 6 to 30) and downstream (uses tones 33 upwards) so tones 31 and 32 appear unused.
Call me 'w23' At any given moment in the universe many things happen. Coincidence is a matter of how close these events are in space, time and relationship. Opinions expressed in forum posts are those of the poster, others may have different views.
on my 20cn 29 to 38 are the gap between up/down and up starts at about 7 thanks davidb ive been having a lot of hr issues and the spectral mask appers to of broken down between up/down my phone has also been cutting of the adsl ,so it seems to fit with it being between them
Let's say you are right and the one guard carrier is not enough. How much bleeding will you get on the other carriers? Let say 2 more carriers are wiped out completely this will reduce the speed by 120kbps
Quote
The QAM rate is said to be 4,000 symbols per second, therefore each 3dB of SNR available in the sub channel over the base line will give approx 4kbps of sync speed, subject to a maximum of 60kbps (15 x 4kbps) per carrier.
Itsme, Aha - I think the DMT graphs on Kitz page may show the issue Matt was referring to. If you look at the DMT graph under ~ Bit Allocation and Power Cut Back. The upstream bits seem to be fully loaded, presumably uncapped upstream. As a result the downstream seems to start later with an upward curve. Now look at the DMT graph under ~ Bit Allocation Table and Low SNR. The upstream bits seem half loaded, presumably capped upload. The downstream seems to start earlier and the bit loading starts high without the upward curve. As you both say - a slight impact on the downstream.
on my 20cn 29 to 38 are the gap between up/down and up starts at about 7 thanks davidb ive been having a lot of hr issues and the spectral mask appers to of broken down between up/down my phone has also been cutting of the adsl ,so it seems to fit with it being between them
Interesting. I'm on 20CN on a very good line. On the DMT tool Upstream is from tone 8 to 30 inclusive Downstream starts at tone 33
